852
The Total Synthesis of Cephalosporin C1
MeOOC
MeOOC
OH
Sir : Cephalosporin C (I), a product of the metabolism of
:n*
H,*H B J O C O N ~ S
COOH
M:-'"M~
V
'HZ OCoCH3
I
COOH
MeOOC
Cephalosporium acremonium, was isolated by Newton and Abraham in 1955,2and its structure was established in 1961 through chemical3 and X-ray crystallographic4 studies. The unusual antimicrobial properties of cephalosporin C and especially of substances prepared from the natural product by chemical modification, of which cephalothin (11)5 may serve as an example, have attracted widespread interest. We now wish to record the first total syntheses of cephalothin and cephalosporin C. L-( +)-Cysteine was converted into L-( -)-2,2-dimethyithiazolidine-4-carboxylic acid,6 which with pyridine and t-butyloxycarbonyl chloride, prepared in situ from t-butyl alcohol, phosgene, and pyridine in methylene chloride at - 74', gave L-( -)-N-t-butyloxycarbonyl-2,2-dimethylthiazolidine-4-carboxylicacid' (111,
,H +
NH2
H&H I BULSOCO N
11
MeOOC H&..H
$OOMe N-NHCOOMe
Y BurocoNvs McMc Me Me
(1) Ihe material presented in this communication provided the basis for the Nobel Lecture delivered by one of us in Stockholm on Dec 11, 1965. (2) G. G. F. Newton and E. P. Abraham, Nature, 175, 548 (1955); Biochem. J., 62, 651 (1956). (3) E. P. Abraham and G. G. F. Newton, ibid., 79, 377 (1961). (4) D . C. Hodgkin and E. N. Maslen, ibid., 79, 393 (1961). (5) R. R. Chauvette, E. H. Flynn, B. G. Jackson, E. R. Lavagnino, R. B. Morin, R. A. Mueller, R. P. Pioch, R . W. Roeske, C. W. Ryan, J. L. Spencer, and E. Van Heyningen, J . Am. Chem. SOC., 84, 3402 ( 1962). (6) G. E. Woodward and E. F. Schroeder, ibid., 59, 1690 (1937). (7) Elemental analytical data in excellent accord with theory were obtained for this substance.
88:4
HO
S
D
B J o c o ~ y sI
H
M e 'Me
Mi -Me
vu
VXI
COOCH2CCC3
IV R = H), mp 114-114.5", [ c Y ] ~ ~-85" D (c 1.34, CHCl,). The corresponding ester7 111 (R = Me), mp 18.5-21", [ a I z 0-77" ~ ( e 1.50, CHCla),prepared from the acid with diazomethane, reacted with excess dimethyl azodicarboxylate at 105" during 45 hr to give the hydrazo diester' Iv, mp 136-137.5', [CY]~'D f 9 8 " ( c 1.45, CHCI,). Oxidation of IV with 2.2 moles of lead tetraacetate in boiling benzene for 2 hr, followed by treatment of the reaction mixture with excess anhydrous sodium acetate in boiling dry methanol for 24 hr, gave the trans hydroxy ester7 V, mp 101-102", [CX]~'D+48"
\ ~
[ a ] " ~-113" (c 0.80, CHCI3), whose structure was confirmed by a complete X-ray crystallographic study.5 With triisobutylaluminum in toluene, the amino ester ~ afforded the @-lactam' VIII, mp 120.5", [ a I Z n-274" ( e 0.52, CHCl,), the structure of which was also confirmed by X-ray crystallographic analysis;* it is worthy of note that this remarkable substance contains the basic structural elements common to cephalosporin C, its transformation products, and the penicillins. In a parallel series of reactions, di-P,P,P-trichloroethyl d-tartrate,' nip 101.5-103.5°, [CY]*~D4-9' (c 1.04, CHCl,), from d-tartaric acid and excess P,P,Ptrichloroethanol in boiling toluene in the presence of p toluenesulfonic acid, was oxidized by sodium metaperiodate in aqueous methanol to @,P,@-trichloroethyl glyoxylate hydrate,' mp 94.5-95.5", which was condensed in aqueous solution with the sodium salt of malondialdehyde to give the aldol7 IX, mp 114-116",
I11
Journal of the American Chemical Society
VI
( e 1.14, CHC13).8 The latter was transformed by treatment in dimethylformamide with excess diisopropylethylamine and methanesulfonyl chloride, followed by concentrated aqueous sodium azide, to the cis azido ester' D (c 1.01, CHCl,), which was VI, mp 55-56", [ c Y ] ~ ~-525" reduced in methanol by aluminum amalgam at - 15" during 24 hr to the cis amino ester7 VII, mp 64-65.5",
-0oc.
ROOC
N3
I
CHOH I
o.., o, H
IX
0
0 X
,A,X,
247 mp ( E 19,100, EtOH-H+), 269 mp (27,600, When IX in 1,2-dimethoxyethane soluEtOH-OH-). tion was added to distilling n-octane, dehydration occurred with formation of the highly reactive dialdehyde X,, , ,A 234 mp (cyclohexane), which was utilized directly, without purification. Condensation of the 6-lactam VI11 with the dialdehyde X in n-octane at 80" during 16 hr afforded the adduct'
(8) The structure of this ester was most convincingly demonstrated by its preparation from the action of diazomethane upon the correspondingacid,'mp 187.5", [ a ] % +123" [c 1.02 (1 N NaOH)], obtained by an alternative method; the structure of the acid was rigorously established by a complete X-ray crystallographic study. This and the other crystal structure determinations mentioned in the sequel were brilliantly executed by Dr. J. Zanos Gougoutas (Harvard), to whom we are glad to express our warm appreciation.
1 February 20, 1966
853 COOCH2CC13
H
0''
& C J ? H o :
=:QCHo
*" ocoNvs NH2 Me- 'Me
C13CCH2OOC fHCH2CH2CH2CO NH
S
YH
XI, [ ~ ] * O D -122" (c 0.46, CHCI,), which in trifluoroacetic acid solution at room temperature during 2.5 hr 292 was transformed to the aminoaldehyde XII, A, mp (13,600, EtOH); this sensitive intermediate was used in subsequent reactions without extensive purification. Condensation of the aminoaldehyde XI1 in benzene with thiophene-2-acetyl chloride in the presence of pyridine led to the amide7 XIII, mp 135-135.5", [aIz0D +485" (c 1.14, CHCL), which was treated in tetrahydrofuran solution with diborane, followed by pyridineacetic anhydride, to give isocephalothin p,p,p-trichloroethyl ester' (XIV), mp 102", [ ~ ] * O D +320" (c 1.01,
coocH2ccI,
&H c-Q H c02-3;
CI,CCH200C C HCH2CH2C%CONH t)H COOCH2CCl3
XIV
9-N
XV [ a I z 0$14" ~ (c 0.95, CHCl3), which was easily sepa-
rated by chromatography on silica gel, and reduced by zinc dust in 90% aqueous acetic acid at room tempera$50" ture to cephalothin (11), mp 160-160.5", [ a ] " ~ (c 1.03, CH3CN), whose properties were identical in all respects with those of material prepared from natural cephalosporin C.6 In another series of reactions the aminoaldehyde XI1 was condensed in tetrahydrofuran with N-P,P,Ptrichloroethyloxycarbonyl-D-(-)-a-aminoadipic acid,? mp 137.5", [ C ~ ] ~ O D -8" (c 1.03, 1 N NaOH), in the presence of dicyclohexylcarbodiimide. The crude reaction mixture was esterified, using P,P,P-trichloroethanol in methylene chloride in the presence of dicyclohexylcarbodiimide and pyridine. Elution of the resulting material from silica gel by benzene-ethyl acetate (3 : 1) gave two products, of which XVI was the more polar, since it was converted by reduction in
CO OC H2CC13
XVII
tetrahydrofuran with diborane, followed by acetylation with acetic anhydride-pyridine to the is0 ester7 XVII, +220" (c 1.01, CHC1,). When the mp 111-1 14", [ C U ] ~ D latter was allowed to stand in pyridine at room temper= ature for 3 days, it was equilibrated (KnormalIiso with the normal ester7 XVIII, mp 157-159", [aI2O~
XI11 qOOCH,CCl3
CHCI,). When the is0 ester was allowed to stand in anhydrous pyridine solution at room temperature during 3 days it was smoothly equilibrated (Knormal,iso = 1/3) with the normal ester7 XV, mp 120-123",
xVI
COOCHfiC13
XI1
YH
COOCH2CClj
XVIII
$40" (c 0.76, CHCL), which was easily separated by chromatography on silica gel and reduced by zinc dust and 90% aqueous acetic acid at 0" during 2.5 hr to cephalosporin C (I), identical with natural material in paper chromatographic behavior, and in antibacterial activity against Neisseria catarrhalis, Alcaligenes faecalis, Staphylococcus aureus, and Bacillus subtilis; further, the synthetic crystalline barium salt, [aI2O~ +80° ( c 0.57, H20), was identical in optical rotation and spectroscopic properties with the salt of natural cephalosporin C. Acknowledgment. We are pleased to acknowledge the very generous support we have received from CIBA, Ltd., and especially the never-failing confidence and encouragement of Dr. Albert Wettstein. R. B. Woodward, K. Heusler J. Gosteli, P. Naegeli, W. Oppolzer R. Ramage, S . Ranganathan, H. Vorbruggen Woodward Research Institute 4000 Basel 7, Switzerland Received January 3, I966
Pyracyloquinone Sir : Although Huckel's 4n 2 rule strictly applies only to monocyclic hydrocarbons, polycyclic aromatics possessing this number of x electrons are frequently misnamed Huckel aromatics. Nevertheless, the simple LCAO calculations do predict these compounds to have a high resonance energy. Brown' predicted that 2 aromatic hydrocarbon, should pyracylene (I), a 4n
+
+
(1)
R. D.Brown, J . Chem. SOC.,2391 (1951). Communications to the Editor
